Conformal prediction for the example from the vignette returns NAs

[ashgreat]

I ran the example from the vignette on multisynth and then generated summary using conformal prediction. However, it returns NAs for standard errors, lower bound, and upper bound:

Note: I made minimal coding changes to the example.

library(augsynth)

data <- read.csv("https://dataverse.harvard.edu/api/access/datafile/:persistentId?persistentId=doi:10.7910/DVN/WGWMAV/3UHTLP", sep="\t")

data |> 
  dplyr::filter(!State %in% c("DC", "WI"),
         year >= 1959, year <= 1997) |> 
  dplyr::mutate(YearCBrequired = ifelse(is.na(YearCBrequired), 
                                 Inf, YearCBrequired),
         cbr = 1 * (year >= YearCBrequired)) -> analysis_df

# with default nu
ppool_syn <- multisynth(lnppexpend ~ cbr, State, year, 
                        analysis_df)

summary(ppool_syn, inf_type = "conformal")
#> 
#> Call:
#> multisynth(form = lnppexpend ~ cbr, unit = State, time = year, 
#>     data = analysis_df)
#> 
#> Average ATT Estimate (Std. Error): -0.011  (NA)
#> 
#> Global L2 Imbalance: 0.003
#> Scaled Global L2 Imbalance: 0.019
#> Percent improvement from uniform global weights: 98.1
#> 
#> Individual L2 Imbalance: 0.028
#> Scaled Individual L2 Imbalance: 0.096
#> Percent improvement from uniform individual weights: 90.4    
#> 
#>  Time Since Treatment   Level     Estimate Std.Error lower_bound upper_bound
#>                     0 Average -0.004280701        NA          NA          NA
#>                     1 Average -0.010855744        NA          NA          NA
#>                     2 Average  0.004379480        NA          NA          NA
#>                     3 Average  0.001156364        NA          NA          NA
#>                     4 Average -0.009304510        NA          NA          NA
#>                     5 Average -0.016942492        NA          NA          NA
#>                     6 Average -0.018504802        NA          NA          NA
#>                     7 Average -0.003866051        NA          NA          NA
#>                     8 Average -0.015834891        NA          NA          NA
#>                     9 Average -0.031750413        NA          NA          NA
#>                    10 Average -0.017838072        NA          NA          NA

^{Created on 2024-08-06 with reprex v2.1.1}

Am I missing something?

[ashgreat]

It looks like conformal prediction is available only for the cases with single treated unit. Can someone please confirm if this is correct?

[ashgreat]

It looks like conformal prediction is available only for the cases with single treated unit. Can someone please confirm if this is correct?

Eli confirmed that currently (as of 08/12/2024) they do not have conformal inference procedure implemented with multiple treated units.